Almost all living organisms have evolved DNA alkyltransferases to repair mutagenic and carcinogenic adducts induced at the O6-position of guanine in DNA by alkylating agents such as methylnitrosourea. Such O6-alkylguanine- DNA alkyltransferase (MGMT previously called GATase) can incidentally repair precursors of cytotoxic DNA crosslinks induced by anticancer chloroethylnitrosoureas (CENUs). All normal human cell lines (Mer+/Mex+) contain MGMT whereas 20-30% of tumor cell lines (Mer-/Mex-) lack this activity and are hypersensitive to the therapeutic effects of these drugs as well as to the mutagenic effects of simple methylating agents. The Mex- subset of tumors would be expected to respond favorable to therapy with CENUs; however, because solid tumors are invariably infiltrated by normal cells, it is not possible to identify Mex- tumors by simply measuring GATase activity in extracts of tumor biopsies. A goal of this proposal therefore is to develop histochemical assays for MGMT expression in human tumor and tissue preparations. To this end we are producing highly specific monoclonal and polyclonal antibodies. Moreover the recent cloning of the cDNA for human MGMT enables us to determine the levels of mRNA and the gene for MGMT in such preparations. Preliminary data suggest that Mex- tumor cells express neither the MGMT protein nor the mRNA hence we hypothesize that regulation of transcription is crucial for the Mex phenotype. We plan to extend these observations to a larger series of other tumor lines including human rhabdomyosarcoma and brain tumor xenograft lines having a spectrum of MGMT expression, as well as normal differentiated human tissues. We will also determine if gene deletion, amplification or rearrangement correlate with MGMT levels in some cases. We will examine transcriptional control mechanisms by determining if differences in MGMT mRNA levels in Mex + and Mex - cells are associated with changes in methylation status of the gene, or with changes in cisacting regulatory elements of the gene, or with changes in trans-acting protein transcription factors. Ultimately we will isolate and characterize such regulatory proteins. Progression of normal diploid cells from Mex+ to Mex- during virally induced transformation and immortalization will be used as a model for investigating MGMT regulation. Detailed analysis of the structure, function, and regulation of the molecular elements of human MGMT activity will ultimately enable one to predict tumor resistance to chloroethylnitrosoureas, as well as an individual's susceptibility to carcinogenesis. It may also indicate where one could successfully intervene in these processes. GRANT=R01CA16265 Our recent work has established three distinct routes for conversion of 7, 12-dimethylbenz(a)anthracene to the carcinogenic precursor 3, 4-dihydrodiol in, respectively, embryo fibroblasts, adrenal cortex, and liver. Two new forms of cytochrome P450 which are very effective in DMBA activation have been identified in mouse embryo fibroblasts (P450-EF) and in rat adrenal cortex (P450-RAP). These cytochrome have been purified, and specific anti- P450 IgGs have been generated that will be used to clone the corresponding cDNAs using libraries generated from, respectively, C3H/10T1/2 mouse embryo fibroblasts and rat adrenal cortex. The relative inductions of P450-EF by benzo(a)anthracene and TCDD suggest regulatory mechanisms in addition to stimulation through the Ah-receptor. The participation of the Ah-receptor will be tested by use of an antagonist and also congenic Ah-responsive and non-responsive embryo fibroblasts. Other mechanism, including protein stabilization, will be tested. P450-EF cDNA and anti-P450-EF IgG will be used to determine the mechanism of induction, and ultimately we aim to obtain 5'-flanking sequences for this gene that will permit definition of the molecular mechanism of regulation in confluent cells and in tumors grown from transformed embryo fibroblasts. Defined serum-free media will be used to determine whether P450-EF is regulated by specific extracellular factors. Specific cDNA, anti-P450 IgG, and antibody-sensitive DMBA metabolism will be used to study the distribution and regulation of P450-EF (in fibroblasts, skin, and mammary cells) and P450-RAP (in adrenal, ovary, and testis) in response to a variety of growth factors and hormones. Specificity for metabolism of different PAHs and potential natural substrates (steroids, eicosanoids, fatty acids) will be examined. In liver, previous data leads us to evaluate two membrane controls over DMBA metabolism: (1) activation of P450IIC6 by phosphatidylserine and cytochrome b5; and (2) selective suppression of DMBA metabolism catalyzed by P450IIB1 relative to P450IIB2 by those or other membrane components. Mechanisms defined for these P450 cytochromes will be assessed for other P450s known to be active in PAH metabolism (including P450-EF, P450-RAP, and P450IA1). Expression of normal and modified P450s by use of transfection techniques will allow molecular interpretation of these changes in DMBA metabolism.